Abstract

     This paper studies the US and global economic fundamentals that exacerbate emerging stock markets volatility and can be considered as systemic risk factors increasing financial stability vulnerabilities. We apply the bivariate HEAVY system of daily and intra-daily volatility equations enriched with powers, leverage, and macro-effects that improve its forecasting accuracy significantly. Our macro-augmented asymmetric power HEAVY model estimates the inflammatory effect of US uncertainty and infectious disease news impact on equities alongside global credit and commodity factors on emerging stock index realized volatility. Our study further demonstrates the power of the economic uncertainty channel, showing that higher US policy uncertainty levels increase the leverage effects and the impact from the common macro-financial proxies on emerging markets’ financial volatility. Lastly, we provide evidence on the crucial role of both financial and health crisis events (the 2008 global financial turmoil and the recent Covid-19 pandemic) in raising markets’ turbulence and amplifying the volatility macro-drivers impact, as well.

Introduction

     A common stylized fact about emerging economies is the high volatility of their stock markets (De Santis 1997; Aggarwal et al. 1999; Xu 1999; Cano-Berlanga and Giménez-Gómez 2018). The liberalization of the emerging world’s financial markets, which attracted a significant amount of capital flows by foreign institutional investors, has been the first step of an integration process with significant economies’ interdependence and asset markets’ synchronization. Given that emerging economies are characterized by critical vulnerabilities to external shocks, they exhibit higher equity market fluctuations than the developed markets and it is worth investigating how US and global common economic forces affect their intra-daily volatility. In general, modeling the volatility of financial returns has crucial implications for asset allocation, risk management practices, and financial stability oversight. Robust modeling and reliable forecasting of the volatility trajectory of financial instruments has been the main task and objective of financial economics applications for business operations, given that volatility constitutes one of the fundamental input variables in estimations and decision processes of any corporation on investing and funding choices. Financial volatility is also closely inspected by policymakers since it is perceived to constitute an early warning crisis signal, entailing critical destabilizing threats for the financial system (see, for example, Kürüm et al. 2018).

Conclusions

     Our study has examined the HEAVY model and its extension with leverage, power transformations, and macro-characteristics. For the realized measure, our empirical results favor the macro-augmented cross asymmetric power specification, where the lags of both powered variables—squared negative returns, and realized variance—move the dynamics of the power transformed conditional variance of the latter. Similarly, modeling the returns with a double asymmetric power process, we found that not only the powered realized measure asymmetry but the power transformed squared negative returns, as well, help to forecast the conditional variance of the latter. The macro-augmentation of the asymmetric power model ensures the superiority of our contribution, which can be implemented in several investment and risk management practices. We further demonstrated the forecasting dominance of the extended specifications over the benchmark HEAVY and standard volatility models through the out-of-sample forecasting across multiple short- and long-term horizons.

     Moreover, we demarcate our study from previous literature by estimating the significant US uncertainty effect on the power of leverage (Heavy and Arch), and the macro-determinants of emerging markets realized variance. The US-led uncertainty spillovers shed light on new evidence for volatility modeling and macro-financial linkages literature. Our findings’ novelty is twofold: Given higher (lower) daily US policy uncertainty levels, mostly associated with economic downturns (upturns), (i) heavy and leverage effects become more (less) acute in realized variance modeling, and (ii) US EPU volatility, financial uncertainty, credit conditions, commodity market benchmarks, and disease news impact on emerging financial volatility increases (decreases). Similarly, financial and health crisis events magnify further the Heavy, Arch, and Macro parameters of the bivariate system and the EPU indirect impact on the volatility drivers, as well.